3. CONTENTS * iii
CONTENTS
ACKNOWLEDGEMENTS.......................................................................................................vi
INTRODUCTION .................................................................................................................1
1 RAPID ASSESSMENT
1.1 Objectives ........................................................................................................5
1.2 Composition of the team ...................................................................................6
1.3 Methods of data collection ...............................................................................6
1.4 Survey and sampling methods ............................................................................8
1.5 Data to be collected .......................................................................................18
1.6 Analysis and presentation of results ................................................................25
2 PREVENTION
2.1 Shelter ...........................................................................................................27
2.2 Water .............................................................................................................33
2.3 Sanitation ......................................................................................................40
2.4 Vector control ................................................................................................46
2.5 Food and nutrition ..........................................................................................64
2.6 Vaccination ....................................................................................................74
2.7 Health education and community participation ................................................88
3 SURVEILLANCE
3.1 General principles ..........................................................................................90
3.2 Objectives ......................................................................................................91
3.3 Setting surveillance priorities .........................................................................91
3.4 Data collection methods .................................................................................93
3.5 Case definitions .............................................................................................94
3.6 Minimum data elements ..................................................................................95
3.7 Data sources for routine surveillance ..............................................................99
3.8 Identifying tasks and responsible persons ......................................................101
3.9 Analysis and interpretation of surveillance data .............................................104
3.10 Feedback .....................................................................................................105
5. CONTENTS * v
ANNEXES
1 WHO reference values for emergencies ..........................................................197
2 Health assessment – sample survey forms ......................................................199
3 NCHS/WHO normalized reference values for weight for height by sex ...............219
4 Weekly surveillance report ............................................................................221
5 Case definitions ...........................................................................................226
6 Outbreak investigation kit .............................................................................232
7 Organization of an isolation centre and calculation of treatment supplies .......236
8 Collection of specimens for laboratory analysis..............................................242
9 Setting up a diagnostic laboratory .................................................................253
10 Treatment guidelines ....................................................................................268
11 Management of the child with cough or difficulty in breathing .......................275
12 Assessment and treatment of diarrhoea .........................................................280
13 Flowcharts for syndromic management of sexually transmitted infections ........285
14 Health card ..................................................................................................288
15 List of WHO guidelines on communicable diseases ..........................................290
16 List of publishers ..........................................................................................293
17 General references .......................................................................................294
6. vi * Communicable disease control in emergencies – A field manual
ACKNOWLEDGEMENTS
Edited by Máire Connolly, WHO/CDS.
Rick Brennan (IRC), Philippe Calain (WHO/CDS), Michelle Gayer (WHO/CDS),
Tim Healing (Merlin), Myriam Henkens (MSF), Jean Long (Trinity College,
Dublin), Serge Male (UNHCR), Pamela Mbabazi (WHO/CDS), Agostino
Paganini (UNICEF), Jean Rigal (MSF), Mike Ryan (WHO/EPR), Peter Salama
(CDC), Paul Spiegel (CDC), Mike Toole (Macfarlane Burnet Centre for Medical
Research and Public Health) and Ron Waldman (Mailman School of Public
Health, Columbia University) contributed extensively to the development of
this manual.
The following people contributed to the development and review of this
document and their input is gratefully acknowledged:
Samira Aboubaker (WHO/CAH), Nathalie Agata (WHO/Ethiopia), Roberta
Andraghetti (WHO/EPR), Ray Arthur (WHO/EPR), O. Babu-Swai (UNHCR,
Kenya), Andrew Ball (WHO/HIV), Claudio Beltramello (WHO/CPE), Sylvie
Briand (WHO/NTD), Nan Buzard (Sphere Project), Claire-Lise Chaignat
(WHO/NTD), Claire Chauvin (WHO/IVB), Denis Coulombier (WHO/EPR),
Charles Delacollette (WHO/RBM), Mike Deppner (UNHCR, Uganda), Philippe
Desjeux (WHO/CPE), Hans Everts (WHO/IVB), Albis Francesco Gabrielli
(WHO/CPE), Bernardus Ganter (WHO/EURO), Antonio Gerbase (WHO/HIV),
Robin Gray (WHO/EDM), Tom Grein (WHO/EPR), Malgosia Grzemska
(WHO/STB), Pierre Guillet (WHO/NTD), Zoheir Hallaj (WHO/EMRO), Max
Hardiman (WHO/EPR), Christopher Haskew (WHO/CPE), Mary Healy
(Trocaire, Ireland), Ana Maria Henao-Restrepo (WHO/IVB), Brad Hersh
(WHO/IVB), David Heymann (WHO/CDS), Gottfried Hirnschall (WHO/HIV),
Jose Hueb (WHO/PHE), Yvan Hutin (WHO/IVB), Jean Jannin (WHO/NTD),
7. ACKNOWLEDGEMENTS * vii
Vijay Kumar (WHO/SEARO), Joël Lagoutte (ICRC, Geneva), Daniel Lavanchy
(WHO/EPR), Dominique Legros (Epicentre), Alessandro Loretti (WHO/HAC),
Paul Lusamba-Dikassa (WHO/AFRO), Chris Maher (WHO/Polio), Frédérique
Marodon (WHO/CPE), Adelheid Marschang (IFRC), Zahra Mirghani
(UNHCR), Lulu Muhe (WHO/CAH), Mike Nathan (WHO/NTD), Maria Neira
(WHO/CPE), Hitoshi Oshitani (WHO/WPRO), Brian Pazvakavambwa
(WHO/HIV), William Perea (WHO/EPR), Pierre Perrin (ICRC), Aafje Rietveld
(WHO/RBM), Guénaël Rodier (WHO/CSR), Cathy Roth (WHO/EPR), Maria
Santamaria (WHO/CSR), Akihiro Seita (WHO/EMRO), Khalid Shibib (WHO/
HAC), Eigil Sorensen (WHO/DPRKorea), John Tabayi (UNHCR, Ethiopia),
Nadia Teleb (WHO/EMRO), Jan Theunissen (WHO/EURO), Michel Thuriaux
(WHO/CSR), A. Tijtsma (UNHCR), Kaat Vandemaele (WHO/EPR), Claude de
Ville de Goyet (WHO/ PAHO), Zita Weise Prinzo (WHO/NHD), Brad Woodruff
(CDC), Nevio Zagaria (WHO/CPE).
WHO would like to thank the Government of Ireland for its support in the
development of this manual.
8. INTRODUCTION * 1
INTRODUCTION
This manual is intended to help health professionals and public health
coordinators working in emergency situations prevent, detect and control
the major communicable diseases encountered by affected populations.
Emergencies include complex emergencies and natural disasters (e.g. floods
and earthquakes). The term “complex emergencies” has been coined to
describe “situations of war or civil strife affecting large civilian populations with
food shortages and population displacement, resulting in excess mortality and
morbidity”.
In this manual, the generic term “emergencies” will be used to encompass all
situations in which large populations are in need of urgent humanitarian relief.
Following an emergency, the affected population is often displaced and
temporarily resettled. They may be placed in camps or become dispersed
among the local population (either in towns or in rural communities). People
who are displaced across national borders are termed refugees whereas those
who have been displaced within their country are called “internally displaced
persons” (IDPs). Resettlement in camps may entail high population densities,
inadequate shelter, poor water supplies and sanitation, and a lack of even basic
health care. In these situations, there is an increased threat of communicable
disease and a high risk of epidemics.
Communicable diseases are a major cause of mortality and morbidity in
emergencies, and particularly in complex emergencies, where collapsing health
services and disease control programmes, poor access to health care,
malnutrition, interrupted supplies and logistics, and poor coordination among
the various agencies providing health care often coexist. The main causes of
morbidity and mortality in emergencies are diarrhoeal diseases, acute respira-
tory infections, measles and, in areas where it is endemic, malaria. Other
communicable diseases, such as epidemic meningococcal disease, tuberculosis,
relapsing fever and typhus, have also caused large epidemics among
emergency-affected populations. Malnutrition and trauma are the two main
additional causes of illness and death.
Ensuring adequate shelter, water, sanitation and food and providing basic
health care are the most effective means of protecting the health of those
affected by emergencies. A systematic approach to the control of communicable
diseases is a key component of humanitarian response, and is crucial to protect
the health of affected populations. This requires co-operation among agencies
working at local, national and international levels, and collaboration among all
sectors involved in the emergency response – health, food and nutrition, shelter,
water and sanitation.
This field manual is the result of collaboration among a number of WHO
departments and several external partner agencies in reviewing existing
9. 2 * Communicable disease control in emergencies – A field manual
guidelines on communicable disease control and adapting them to emergency
situations. This manual deals with the fundamental principles of communicable
disease control in emergencies, which are:
• RAPID ASSESSMENT: identify the communicable disease threats faced
by the emergency-affected population, including those with epidemic
potential, and define the health status of the population, by conducting a
rapid assessment;
• PREVENTION: prevent communicable disease by maintaining a healthy
physical environment and good general living conditions;
• SURVEILLANCE: set up or strengthen disease surveillance system with
an early warning mechanism to ensure the early reporting of cases, to monitor
disease trends, and to facilitate prompt detection and response to outbreaks;
• OUTBREAK CONTROL: ensure outbreaks are rapidly detected and
controlled through adequate preparedness (i.e. stockpiles, standard
treatment protocols and staff training) and rapid response (i.e. confirmation,
investigation and implementation of control measures); and
• DISEASE MANAGEMENT: diagnose and treat cases promptly with
trained staff using effective treatment and standard protocols at all health
facilities.
It is hoped that this manual, by setting standards for communicable disease
control in emergencies, will promote effective, coordinated action towards the
prevention and control of communicable diseases in emergencies. Where
appropriate, the manual provides suggestions for further reading, with
references to relevant background material, guidelines and reviews. Finally,
whilst shelter, food, water and sanitation sectors are covered, this manual
specifically aims to provide detail on health issues.
10. INTRODUCTION * 3
Steps in ensuring communicable disease control in emergencies
Conduct rapid health assessment
• Identify main disease threats, including potential epidemic diseases
• Obtain data on the host country, on the country of origin of displaced persons
and on the areas through which they may have passed
• Identify priority public health interventions
• Identify the lead health agency
• Establish health coordination mechanisms
➜
Prevent communicable diseases
• Select and plan sites
• Ensure adequate water and sanitation facilities
• Ensure availability of food
• Control vectors
• Implement vaccination campaigns (e.g. measles)
• Provide essential clinical services
• Provide basic laboratory facilities
➜
Set up surveillance/early warning system
• Detect outbreaks early
• Report diseases of epidemic potential immediately
• Monitor disease trends
➜
Control outbreaks
Preparation – outbreak response team
– stockpiles
– laboratory support
– standard treatment protocols
Detection – surveillance/early warning system
Confirmation – laboratory tests
Response – investigation
– control measures
Evaluation
11. 1. RAPID ASSESSMENT * 5
1. RAPID ASSESSMENT
A rapid health assessment must be conducted as soon as possible after an
emergency, ideally within one week. The aim is to identify the main commu-
nicable disease threats, outline the public health needs and plan priority
interventions. The duration of a rapid assessment depends on the size and
geographical distribution of the population affected, the security situation, the
conditions of access, transport and logistics, the human resources available
and the methods used. It should be completed within one week, depending on
the extent of the emergency.
A more thorough assessment, with detailed qualitative and quantitative data
and intervention plans, should be completed as soon as possible after the rapid
assessment. The assessment must be undertaken by well-qualified and
experienced epidemiologists. The key activities involved in a rapid assessment
are outlined in Table 1.1.
Table 1.1 Key activities in rapid assessment
1. Planning the mission
Composition of the health assessment team
Collection of background geopolitical data
Collection of background health data on host country and country of origin
2. Field visit
Data: demography, environment, health data, resource needs
Methods: aerial inspection; direct observation; interviews with agencies, the ministry of health and
local authorities; collection of health data from medical facilities; rapid estimation of population size
by mapping, review of records and rapid surveys
3. Analysis
Demographic pyramids
Priority health interventions
Identification of high-risk groups
4. Report writing
5. Dissemination
1.1 Objectives
The objectives of a rapid assessment are:
• to assess the extent of the emergency and the communicable disease threat
to the population;
• to define the type and size of interventions and priority activities;
• to plan the implementation of these activities;
• to pass information to the international community, donors and the media
in order to mobilize human and financial resources.
12. 6 * Communicable disease control in emergencies – A field manual
1.2 Composition of the team
The rapid health assessment team should consist of:
• a public health expert/epidemiologist,
• a nutritionist,
• a logistician/administrative officer,
• a water and sanitation/environmental health specialist.
One member must be designated as team leader.
The tasks of the team during this initial phase are to:
• prepare a rapid health assessment checklist,
• prepare a timetable of assessment,
• assign tasks,
• obtain necessary equipment (e.g. computers, scales, laboratory supplies),
• organize visas, transport, vehicles, fuel,
• set up a communication system,
• inform the local authorities,
• inform potential donors and key decision-makers.
1.3 Methods of data collection
The collection of data in an emergency may not proceed in a step-by-step
manner, but the plan for data collection and analysis must be systematic.
The four main methods of collecting data are:
• review of existing information,
• visual inspection of the affected area,
• interviews with key informants,
• rapid surveys.
1.3.1 Review of existing information
A review should be undertaken of baseline health and other information
available at national and regional levels, from government, international and
nongovernmental sources, concerning:
• the geographical and environmental characteristics of the country and
affected area (e.g. national, subnational and district maps showing admin-
istrative and political divisions of the affected area, settlements, water
sources, main transport routes and health facilities);
• the size, composition and prior health and nutritional condition of the
emergency-affected population;
• the health services and programmes functioning before and during
the emergency;
13. 1. RAPID ASSESSMENT * 7
• the resources already allocated, obtained or requested for the emergency
response operation;
• the security situation.
1.3.2 Visual inspection of the affected area
Where travel is by air, useful observations of the affected area can be made before
landing. An initial walk or drive through the area may allow for a first rough idea
of the adequacy of shelter, food availability, environmental factors such as
drainage and risks of vector breeding, and the general status of the population.
During the initial visual inspection, the area should be mapped, even if only
crudely. The resulting maps should indicate the affected area, the distribution of
the population and the location of resources (medical facilities, water sources,
food distribution points, temporary shelters, etc.).
Mapping also allows the estimation of population data, through the calculation
of the total surface area of the camp and of sections of the camps. The method
is based on the making of a map of the camp, with its different sections. By using
random sampling of several known surface areas one may count the number of
persons living in these zones and establish the average population per area. One
can calculate the total population of the camp by extrapolating the average
population per square into the total surface area of the site (see example).
1.3.3 Interviews with key informants
Interviews must be conducted with key personnel in the area and with the
affected population, and must include people from all sectors of the population
involved:
• clan, village and community leaders,
• area administrators or other governmental officials,
• health workers, including traditional birth attendants, healers, etc.,
• personnel from local and international emergency response organizations,
including United Nations organizations,
• individuals in the affected population.
Community organizational structures, normal dietary practices, cultural practices
relating to water and hygiene, and preferences for health care should all be
recorded.
1.3.4 Rapid surveys
Rapid surveys take time and should be reserved for essential data not available
from other sources. They may be used to determine the sex and age distribution
of the population, the average family size, the number of people in vulnerable
groups, recent mortality rates (retrospective mortality study), the main causes
of mortality and morbidity, current nutritional status, vaccination coverage, and
14. 8 * Communicable disease control in emergencies – A field manual
the use of formal and informal health services. Techniques are rather well
codified, and are based on validated sampling and analysis methods in order
to provide quantitative estimates of the situation with reasonable accuracy and
within acceptable delays. This is essential (a) to guide emergency decisions on
where and when resources should be allocated, and (b) as a baseline for
monitoring interventions. Surveys and sampling methods are outlined below.
Sample household survey and rapid assessment forms are included in Annex 2.
1.4 Survey and sampling methods
1.4.1 Introduction
The first priority when entering an emergency area for the first time is to
undertake a needs assessment so as to ensure an effective use of limited
resources. Inadequate or incomplete assessments can lead to inappropriate
responses and waste of scarce resources, and personnel may be needlessly
endangered.
While most assessments will be a straightforward data collection exercise, a
structured and statistically analysable survey may be needed to answer a
particular question. It is equally important to undertake re-surveys at intervals,
so as to keep abreast of a changing situation. This forms an important part of
ongoing surveillance. The use of a standard method throughout means that
the results of different surveys/assessments can be compared directly. Any
changes quickly become apparent. The use of such methods makes it easier to
monitor the response and determine its effectiveness.
If possible, advice should be taken from a biometrician before formal surveys
are undertaken, as it is important to structure the survey so as to get
representative and easily analysable results.
When possible use EpiInfo and EpiData for all aspects of data entry and
handling.
It is important to understand that there may be security implications of
undertaking surveys and assessments in chaotic, unstable situations. Local
people may view questionnaires with suspicion: they may not understand the
idea of the survey and feel that (for example) they are being earmarked for
deportation. Some form of advance publicity may be necessary, but should be
undertaken carefully so as not to bias any samples. The survey/assessment
may need to be undertaken rapidly if the situation is dangerous.
In addition to the rapid needs assessment, good epidemiological surveillance
should be put in place as soon as possible.
Rapid needs assessment should not be viewed in isolation but as one aspect of
surveillance in emergency situations.
Attempts should also be made to:
15. 1. RAPID ASSESSMENT * 9
• secure data from other sources (e.g. clinic data, local authorities, other
nongovernmental organizations, community leaders, etc.);
• ensure that there is ongoing surveillance, possibly using sentinel surveil-
lance points, repeat surveys;
• regularly analyse existing health clinic data, etc.
1.4.2 Surveys
Although the ideal would be to measure the whole of a population, this is rarely
possible (it may occasionally be so in a small refugee camp). In practice a sample
must be taken. The sample should be representative of the population, but a
balance must be struck between the ideal and the attainable. In emergencies
there is always a trade-off between rapidity and accuracy.
The size of the sample must be adequate to accomplish what is required, but
not wasteful. In an emergency the size of a sample may be governed by factors
other than immediate statistical requirements (e.g. accessibility, staff
availability, security, etc.).
There are several essential steps to be followed in any survey:
1. Defining the aims clearly
This is the crucial first step, as all other aspects of the survey stem from this.
Most surveys have multiple aims. The key reasons for an agency to undertake
a survey are to ensure that the appropriate aid is sent to those who need it in
the acute phase of the emergency, and to have a baseline from which to monitor
the effect.
It is important not to try to collect too many items of data in one survey. Define
what you need to know, not what you would like to know. Consider cost, speed,
available resources and security.
2. Selecting the site
You need to decide which area you want to have information about. This could
be, for example, a province, a city, the area where an agency is active or a
damaged area of a city (if a large part was untouched). The area selected should
be clearly defined, together with the reasons for its selection. A suitable control
area may need to be selected. If you are working in a devastated area, an
untouched area might be needed at the same time for comparison.
3. Defining the basic sampling unit
In random sampling methods the basic sampling unit is usually individuals,
whereas in a cluster survey it is usually occupied households. Whatever
definition is chosen, it should be stated in the report.
16. 10 * Communicable disease control in emergencies – A field manual
4. Sample size
The size of the sample should ideally be based on how reliable the final
estimates must be. A sample must represent the population as a whole. Each
individual should have an equal chance of being sampled, and the selection of
an individual should be independent of the selection of any other.
➜ For calculating the sample size for a proportion:
The following items determine the size of the sample needed.
✔ The confidence level and precision required
The use of a sample means that only an estimate of the results from the whole
population is obtained. Subsequent samples are likely to give different values,
but provided the samples have been selected correctly there will be little
variation between them. The actual population value will lie in a range around
the observed value(s). The confidence interval is the upper and lower limits of
this range (e.g. result = 12% ± 2%; the confidence interval is 10–14%). The size
of the confidence interval is related to the error risk and the sample size. The
greater the precision required, the larger the sample must be.
✔ The variability of the characteristics being measured in the study
population
If this is unknown you must assume maximum variability.
✔ The size of the population under study
Usually what is required is to find out what proportion of a population has some
characteristic. When this type of information is sought, sample size is
determined using the following formula:
t 2× p × q
n = ————
d2
where: n = first estimate of sample size
t = confidence (for 95% use 1.96)
d = precision (usually 0.05 or 0.10)
p = proportion of the target population with the characteristics being
measured (if proportion is unknown, let p = 0.5)
q=1–p
Once n is calculated, compare it with the size of the target population (N). If n
is known (or strongly suspected) to be less than 10% of N, then use n as the
final sample size. If n >10% N, then use the following correction formula to
recalculate the final sample size (nf). (If the sample size is >10% of the
population a smaller sample can be used.)
n
nf = ———
1 – n/N
17. 1. RAPID ASSESSMENT * 11
Once nf has been determined, you need to decide whether it is possible to
achieve that sample size under the circumstances of the mission. If not, a
smaller sample size may have to be accepted with the caveat that this will
reduce precision.
As an example, we have a population where the expected disease rate is 12%.
We need to measure the prevalence with a precision of 2%. The sample size
required is:
1.962 × 0.12 × 0.88
n = ———————— = 1014
0.022
However, the population size is known to be 6000 and this sample size is
therefore >10% of the population size.
1014
revised n =—————— = 867
1 – 1014/6000
To assist you, some sample sizes with an error risk of 5% are given in Table 1.2.
(NB. These are not revised sample sizes.)
Table 1.2 Sample size according to expected prevalence in a random or
systematic sample with an error risk of 5%
Prevalence Precision
1% 2% 3% 4% 5%
5% 1825 456 203 – –
6% 2167 542 241 135 –
7% 2501 625 278 156 100
8% 2827 707 314 177 113
9% 3146 787 350 197 126
10% 3457 864 384 216 138
11% 3761 940 418 235 150
12% 4057 1014 451 254 162
13% 4345 1086 483 272 174
14% 4625 1156 514 289 185
15% 4898 1225 544 306 196
16% 5163 1291 574 323 207
17% 5420 1355 602 339 217
18% 5670 1418 630 354 227
19% 5912 1478 657 370 236
20% 6147 1537 683 384 246
30% 8067 2017 896 504 323
40% 9220 2305 1024 576 369
50% 9604 2401 1067 600 384
18. 12 * Communicable disease control in emergencies – A field manual
➜ For calculating the sample size for a mean, for example the population mean:
When sample data is collected and the sample mean χ is calculated, that
sample mean is typically different from the true population mean µ . This
difference between the sample and population means can be thought of as an
error. The margin of error Ε is the maximum difference between the observed
sample mean χ and the true value of the population mean µ :
σ
Ε = z2/α × —–
√ n
where: zα/2 is the “critical value”, the positive value z that is at the vertical
boundary for the area of α/2 in the right tail of the standard normal
distribution
σ is the population standard deviation
n is the sample size
Rearranging this formula, the sample size necessary can be calculated to give
results accurate to a specified confidence interval and margin of error.
z × σ 2
n= 2/α
Ε
This formula is used when you know σ and want to determine the sample size
necessary to establish the mean value µ , with a confidence of (1 – α ), within ±Ε
of error. This formula can still be used if the population standard deviation σ
is unknown and you have a small sample size. Although it is unlikely that the
standard deviation σ is known when the population mean is not known, σ may
be determined from a similar process or from a pilot test/simulation.
5. The interview/questionnaire
Only collect information that will be used. Keep questions simple and unam-
biguous with yes/no answers as often as possible. Keep as short as possible to
save time. Teams should be well trained and not allowed to introduce personal
bias into the sampling. The people chosen should be acceptable and non-
intimidating to the general population.
You may wish to build checks into the questionnaire. For example, two
questions may be included at different places in the questionnaire that are
differently worded but whose answers are the same in whole or in part. If very
different answers are received, the veracity of the respondent may be
questionable.
It may be necessary to select times of day for interviews when people are likely
to be available.
19. 1. RAPID ASSESSMENT * 13
1.4.3 Sampling methods
The following methods are discussed:
1. Census
2. Simple random sampling
3. Systematic sampling
4. Stratified sampling
5. Cluster sampling
1. Census
A census involves determining the size of a population and (often) obtaining
other data at the same time. All the individuals (or at least representatives of
all individuals, such as heads of families) need to be interviewed. This may be
useful in well-defined populations such as refugee camps but can be extremely
time consuming. Registration of refugees is notoriously unreliable, especially
where food is concerned. The situation in acute emergencies is often chaotic
and dynamic. Previous census data are often meaningless owing to massive
population movements. Census data may be required to determine parameters
such as rates of infection in well-defined populations (such as refugee camps),
but such data can usually be obtained from the agency running the camp.
2. Simple random sampling
Random sampling is the only way of meeting the two criteria: that each
individual should have an equal chance of being sampled, and the selection of
an individual should be independent of the selection of any other. The
individuals to be questioned (the “sampling units”) are selected purely by
chance from a complete list of the entire population being studied. (For
example, each individual can be given a number and then numbers selected
from the total list by use of random number tables.)
This is a statistically reliable method but can be time consuming and requires
an accurate list of the individuals in the area. Determining the appropriate size
of the sample can be a problem. The list of individuals could come from refugee
registers, census data, tax registers, electoral registers, etc. In a war with
shifting populations such reliable data rarely exist and this limits the use of this
method. It may be appropriate in refugee camps with good registration data.
No control over the distribution of the sample is exercised, so some samples
may be unrepresentative.
There are a number of specialized techniques, based on random sampling, that
are designed to ensure representative samples.
20. 14 * Communicable disease control in emergencies – A field manual
Tables of random numbers can be generated by Excel in the following way:
– launch Excel
– enter the following formula in the top left corner cell of the table that you want to
produce: =RAND()*100
– you can increase the range of values returned by the formula by adjusting the value 100
in the formula. If 100 is replaced by n, random numbers from 0 to (n –1) will be produced.
– set the decimals to 0 by using the “Format” and “Cell” menu options, and “Number”
and “Decimal places” tabs
– copy the cell containing the formula to the range of the table to be produced
Tables of random numbers can be generated by Epi Info in the following way:
– run EPED from the EPI6 menu
– make a questionnaire file containing only one field: Number ######
– run ENTER and make a .REC file called RANDOM but do not enter any records
– press <F10> to return to the EPI6 menu
– run ANALYSIS from the EPI6 menu and READ RANDOM
– set up the random integers:
NUMBER = RAN(100)
GENERATE 1000
The integers will be placed in the file. You can now use the commands LIST
NUMBER, FREQ NUMBER or LINE NUMBER to check the distribution. To print
them for use, use ROUTE PRINTER followed by LIST NUMBER. If you want real
numbers instead of integers, the RND(100) command will produce these.
21. 1. RAPID ASSESSMENT * 15
3. Systematic sampling
This method is used when individuals or households (sampling units) can be
ordered or listed in some manner. Rather than selecting all subjects randomly,
a selection interval is determined (e.g. every fifth individual) a starting point
on the list picked at random and every nth person, household, etc. is selected
(where n = the sampling interval) on the list. Good geographical distribution
(according to population density) can be assured.
However, sampling units do not need to be listed in emergency situations as
long as:
– the total number of units can be estimated,
– the enumerators can go through the area and pass in front of each sampling
unit, selecting every xth unit.
Systematic sampling allows better representativeness than simple random
sampling (assuming there is no cyclic pattern in the distribution of sampling
units and which would be extremely rare).
Example – Systematic sampling
Number of households: 6000
Sample size: 450 households
Sample interval: 6000/450 = 13 households
Information is collected every 13 households if the departure number is chosen at
random between 1 and the sample interval.
If for example the departure number is 5 (thus the fifth household beginning at one
extremity of the camp), then the selected households are numbers:
5, then 18 (5+13); then 31 (18+13); then 44 (31+13), etc.
4. Stratified sampling
In this method the target population is divided into suitable, non-overlapping
subpopulations (strata). Each stratum should be homogeneous within and
heterogeneous between strata. A random sample is then selected within each
stratum. Separate estimates can be obtained from each stratum, and an overall
estimate obtained for the whole population defined by the strata. The value of
this technique is that each stratum is accurately represented and overall
sampling error is reduced.
5. Cluster sampling
One of the difficulties faced in most disasters is that the size of a population
may not be known. The EPI (Extended Programme on Vaccination) cluster
sampling method is often used to overcome this difficulty, as it is reliable,
relatively cheap and rapid. It was originally developed to assess levels of
22. 16 * Communicable disease control in emergencies – A field manual
smallpox coverage, and has been extended for use in other vaccination
programmes. It is designed to produce representative samples even if the
population size is unknown. Both qualitative and quantitative data can be
collected. This method has been used in emergency situations.
Cluster sampling methods are also valuable when a population is geo-
graphically dispersed. The units sampled first are not members of the
population but clusters (aggregates) of the population. The clusters are selected
in such a fashion that they are representative of the population as a whole. For
example, in a rural area a sample of villages may be selected and then some
or all of the households included in the sample. A major advantage of this
approach is that there is a saving of resources (reduced travelling, fewer staff)
but the method lacks precision when compared to random sampling.
The EPI technique has been modified for use in nutritional surveys. A further
development for the rapid assessment of health needs in disasters uses the
technique to assess multiple aims, and consequently the basic sampling unit
is no longer the individual but the household.
A variant of this method, which has been used in emergency situations, is the
Modified two-stage cluster sampling method.
It is important to realize the limitations of this survey technique.
• The simple 30 × 7 “EPI” design is adequate for relatively frequent events
(e.g. looking at vaccination status in the under-5s (when your sampling unit
would be children under 5) but will not provide accurate estimates of
relatively rare events such as mortality. To do this a much larger sample size
would be needed.
• The technique does not meet the second criterion for representativeness,
because the selection of an individual in a cluster is not independent of the
selection of other individuals. Members of a cluster are likely to be similar.
This is known as the “design effect” and may be especially serious with
communicable diseases owing to their tendency to cluster close together. If
questions about mortality and specific diseases are asked then great care
should be taken not to draw too many firm conclusions from the results.
Cluster sampling is more suited to questions related to, for example, access
to health care, people currently ill, or need and availability of medication. The
design effect can be countered to some degree by doubling the size of the
sample required in random or systematic sampling. Design effect as such
does not affect the point estimate calculated on the sample, but the precision
(variance) only. The decrease in precision can be calculated during analysis
by comparing the variance between clusters over the global variance.
The use of this technique in emergencies needs further rigorous evaluation,
but in the mean time it seems to be the best method for data collection in urban
areas that have been devastated by war or natural disaster.
23. 1. RAPID ASSESSMENT * 17
The EPI method usually samples 30 clusters from the area of interest and then
7 subjects from adjacent households. (In principle the more clusters the better
the randomness, but it should be balanced against time and cost.)
An example of cluster sampling is given below. The population of different
sections in a settlement must be known. This method is divided into two stages
as described below.
Stage 1: Selection of clusters
• Calculate cumulative population total (for example, population = 9300)
• Calculate the sampling interval
For a cluster sampling survey, a sample of 30 clusters of 30 households is
recommended (representing approximately 4000 to 5000 persons, of which
900 are children between 6 and 59 months)
The desired sample size is, therefore, 30 clusters of 30 households =
900 households
The sample interval for selecting the clusters is then calculated:
e.g. 9300/30 = 310
• Determine the first cluster by drawing, at random, a number between 1 and the
sample interval (example in table = 170)
The other clusters are then positioned in the cumulative list (170 + 310 = 480,
etc.) and a number of clusters per section derived
Population per section Cumulative total Number of clusters per section #
1 = 500 500 (170, 480) 2
2 = 2000 2500 (790, 1100, 1410, 1720, 2030, 2340) 6
3 = 1500 4000 (2650, 2960, 3270, 3580, 3890) 5
4 = 1000 5000 (4200, 4510, 4820) 3
5 = 1300 6300 (5130, 5440, 5750, 6060) 4
(6370, 6680, 6990, 7300, 7610,
6 = 3000 9300 10
7920, 8230, 8540, 8850, 9160)
Population total 9300
24. 18 * Communicable disease control in emergencies – A field manual
Stage 2: Selection of individuals
• A direction is chosen at random from the centre of each section, and the number
of households counted from the centre to the periphery in this direction.
• A number between 1 and the number of households counted is chosen at random.
This number corresponds to the household that is the departure point for the
selection of individuals in the cluster.
• In each cluster, households are selected by moving from one household to the
nearest household, until the required cluster size is obtained (a target of 30
households per cluster is recommended).
• When children from 6 to 59 months are the target of the survey and there are
several of these children in one household, one is chosen at random from each
household.
If several clusters have been selected in one section, the same operation is repeated
from the centre of the section.
Note: When possible, systematic sampling should be chosen by preference over cluster sampling. It is easier to
carry out and, above all, is more rapid (one is able to obtain a precision of results equivalent to cluster sampling,
but with a much smaller sample size).
1.5 Data to be collected
Data should be collected in the following areas:
• background health information,
• demography,
• mortality,
• morbidity,
• health services and infrastructure,
• food,
• nutritional status,
• water,
• sanitation,
• shelter and non-food items,
• environment,
• coordination.
1.5.1 Background health information
Background health information comprises:
• the main health and nutritional problems,
• coverage by public health programmes (e.g. vaccination coverage rates),
25. 1. RAPID ASSESSMENT * 19
• the health care infrastructure, staff available and use of traditional medicine,
• the availability of health workers,
• important health beliefs and traditions,
• social organization.
When displaced populations are at the centre of the humanitarian emergency,
similar data should be collected on their place of origin.
If possible, background information should be collected before the field mission
is conducted, using as sources the Ministry of Health, WHO, international and
nongovernmental organizations and WHO web sites.
In the field, data should be collected through interviews with community
leaders, heads of households, health workers and individuals.
1.5.2 Demography
Demographic information comprises:
• total population size (displaced persons and host population);
• population under 5 years of age;
• ethnic composition and place of origin;
• sex ratio;
• the number of persons in the following high-risk groups: pregnant and
lactating women, members of households headed by a woman, unaccom-
panied children, disabled and wounded people, and the elderly;
• the average size of a family/household;
• the number of arrivals and departures per week;
• the predicted number of future arrivals;
• activity patterns in the host (and possibly the displaced) populations that
may affect the timing of surveys (e.g. when people go to collect water, to the
fields, etc.).
The following sources can be used for retrieving demographic data:
• mapping;
• aerial or satellite pictures;
• census data;
• records maintained by camp administrators, local government officials,
United Nations organizations, religious leaders, etc.;
• interviews with leaders among displaced groups.
Survey questionnaires in sampled dwellings should include the number, age
and sex of family members and the number of pregnant and lactating women.
The average number of persons per dwelling visited and the total number of
dwellings in the camp or settlement should be calculated.
26. 20 * Communicable disease control in emergencies – A field manual
1.5.3 Mortality
During a rapid initial assessment, and before any surveillance system can be
put in place, any mortality data will of necessity be retrospective. The choice
of the retrospective time period used to calculate mortality rates will depend
on which critical event(s) influencing mortality have to be included in the survey
estimate. It will also depend on cultural events that stand out in the memories
of those interviewed. A balance must be struck between expectations of greater
precision (requiring longer recall periods) and avoidance of recall bias.
The survey questionnaire should in any case capture, in a culturally sensitive
way, the following:
• total deaths for given period (e.g. one week),
• deaths among those under 5 years of age for the same period,
• major causes of death.
Approximate daily death rates should be calculated daily or weekly, depending
on the severity of the emergency. In the acute phase of an emergency, daily
deaths rates should be calculated as follows:
• crude mortality rate: number of deaths per 10 000 people daily or weekly,
• age-specific mortality rates: number of deaths per 10 000 people in the under-5
and 5-and-over age groups daily or weekly,
• cause-specific mortality rates: number of deaths from a given cause per
10 000 people daily or weekly.
Table 1.3 Thresholds and calculations
Thresholds Calculations
Crude mortality rate D 10000
X
> 1/10 000/day: severe situation (N+D) + N sp
> 2/10 000/day: critical situation
2
Under 5 mortality rate: double (N+D) + N
average population =
> 2/10 000/day: severe situation 2
> 4/10 000/day: critical situation D = number of deaths during the study period
Normal and stable situation: N = number of people of the sample living at the
Developing countries: 0.6/10 000/day end of the study period
Industrialised countries: 0.3/10 000/day sp = study period expressed in days
The following methods can be used to collect mortality data.
• Count the number of graves: designate a single burial site for the camp or
settlement monitored by grave-watchers 24 hours a day, and develop a verbal
autopsy procedure for expected causes of death using standard forms.
• Check hospital/health facility records and records of organizations respon-
sible for burial.
• Interview community leaders.
27. 1. RAPID ASSESSMENT * 21
• For the collection of prospective mortality data, other methods can be used,
such as mandatory registration of deaths, issuing of shrouds to families of the
deceased to help ensure compliance, or employing volunteer community in-
formants to report deaths for a defined section of the population (e.g. 50 families).
1.5.4 Morbidity
The number of cases of disease should include:
• diseases that cause substantial morbidity (i.e. diarrhoea, respiratory
infections and malaria where prevalent);
• diseases that have the potential to cause epidemics (i.e. measles, cholera,
meningitis and haemorrhagic fevers).
Classical sources of morbidity data are:
• patient registers and records in camp or settlement clinics, hospitals or
feeding centres;
• interviews with health workers, midwives within the displaced population;
• records of local hospitals or clinics.
After the acute phase is over, a properly designed emergency surveillance
system should provide more accurate morbidity data (see Chapter 3).
1.5.5 Health services and infrastructure
Access
• Access by the affected population to local, pre-existing health services.
• Ability of local health services to absorb the influx of people affected by the
emergency.
Facilities
• Numbers, names and types of health facilities available, i.e. clinics, hospitals,
feeding centres and laboratories.
• Level of support – ministry of health or nongovernmental organization.
• Level of functioning.
• Level of damage.
• Number of beds including maternity beds – total and occupied currently.
• Average number of outpatients seen per day – 6 months ago and current.
• Average number of deliveries during one week – 6 months ago and current.
• Availability of operating theatres.
• Numbers, type, size and capacity of health facilities set up for the displaced
population if separate (e.g. tent, local materials).
• Adequacy of water supply, vaccine cold chain (freezers and refrigerators),
generators or town electricity, toilets and waste disposal facilities and food
for patients or malnourished.
28. 22 * Communicable disease control in emergencies – A field manual
Health personnel
• Per health facility above, types and numbers of health personnel and relevant
skills and experience present in the hosting area – 6 months ago and current.
• Health workers present among the displaced population, including
traditional healers, traditional midwives, doctors and nurses, laboratory
technicians, and water and sanitation engineers.
• Availability of interpreters.
Drug and vaccine supplies
• Availability of essential drugs and medical supplies (see Annex 2 – Rapid
Health Assessment, Table 1).
• Availability of the WHO New Emergency Health Kit, which contains drugs
and medical supplies for 10 000 people for approximately 3 months (see
Annex 10).
• Availability of essential vaccines and vaccination equipment (e.g. measles
vaccines, injection material and cold chain equipment).
1.5.6 Food
• Number of calories available per person per day.
• Frequency of distribution of food rations.
• Length of time these rations have been provided.
• Food basket monitoring.
Sources of data
– Assessment of the quality and type of food available to the population.
– Inspection of local markets for food availability and prices.
– Assessment of local, regional and national markets for availability of
appropriate emergency foods.
1.5.7 Nutritional status
(See also Section 2.5.)
• Prevalence of acute malnutrition in children 6–59 months of age or 60–110 cm
in height.
• Percentage of children severely and moderately malnourished.
• Prevalence of clinically observable micronutrient deficiencies.
• Feeding programmes currently being planned – number of children being
cared for daily in supplementary feeding programmes (SFP) and therapeutic
feeding programmes (TFP).
• Number of additional calories per day provided by SFP .
Sources of data
– Unbiased representative sampling.
– Mass screening (all children weighed and measured).
29. 1. RAPID ASSESSMENT * 23
1.5.8 Water
• Litres of water per person per day.
• Length of time this quantity has been available.
• Source and quality of water.
• Number and type of water points.
• Water storage facilities.
• Water purification methods available/in use.
• Length of time persons must wait for water.
• Number of persons per water point.
• Transport and storage.
• Equipment/expertise on site, planned or available if needed.
1.5.9 Sanitation
• Current facilities for excreta disposal and population per latrine or toilet.
• Anal cleansing methods and availability.
• Availability of soap.
• Presence of vectors (arthropods, mammals).
• Adequacy of burial sites.
1.5.10 Shelter and non-food items
• Blankets, clothing and domestic utensils.
• Shelter.
• Livestock.
1.5.11 Environment
• Climate.
• Topography and drainage.
• Suitability as site for settlement from health point of view.
• Access (routes to site, road surface, airfields, security issues).
• Transport.
• Amount of land: persons per square metre.
• Building materials.
• Fuel availability.
• Storage facilities for food, medical supplies.
• Communication.
Sources of data
– This assessment is largely carried out by visual inspection. Information must
also be obtained from key informants such as local officials and United
Nations, international and nongovernmental organizations.
30. 24 * Communicable disease control in emergencies – A field manual
– Focus group discussions with the community may also be useful, addressing
such issues as their cultural perceptions of water and sanitation, how they
bury the dead, and where they find food, fuel and shelter materials.
1.5.12 Coordination
The following information should be obtained from national, United Nations,
international and nongovernmental organizations working on the affected area.
• What is the existing local response capacity?
• What is the presence and activities of international or local organizations?
• Who is in charge of coordinating health, water and sanitation activities?
• Who supplies which services in these sectors?
• Who coordinates food delivery to the area and its distribution to the affected
populations?
• What have they achieved to date?
• What are the additional needs in terms of financial and material resources,
and of implementation capacity?
• What are the priorities for immediate action?
A summary of the essential information to be collected during a rapid
assessment is given in Table 1.4.
1.5.13 Common sources of error
Logistic
• Insufficient transport and/or fuel.
• Visas/security clearance not received in time.
• Inadequate communication between field, regional and national levels:
authorities in charge not informed in time and not ready to assist.
Organizational
• Lead organization not identified, team leader not identified, responsibilities
of various organizations not well defined.
• Key decision-makers and donors not informed that an assessment is being
undertaken.
• Assessment conducted too late or takes too long.
• Information collected that is not needed for the emergency response.
Technical
• Specialists with appropriate skills not involved in the assessment.
• Programmes that could be implemented immediately unnecessarily delayed
because of the assessment (e.g. measles vaccination).
• Assessment conclusions not representative of the affected population.
• Surveillance system developed too slowly, thus preventing monitoring and
evaluation of emergency response programme.
31. 1. RAPID ASSESSMENT * 25
Table 1.4 Essential information to be collected during a rapid assessment
• Background to the emergency
• Estimate of size of affected population and population movements
• Map of the site
• Environmental conditions
• Security conditions
• Health and nutritional status of the population affected by the emergency
• Major health threats – communicable and noncommunicable diseases
• Diseases of epidemic potential
• Existing health facilities and staff – capacity to deal with the affected population
• Estimation of recent mortality rates
• Surveillance system in place prior to the emergency
• Availability of food and water
• Extent of involvement of the local authorities, especially the Ministry of Health
• Presence and activities of international or local organizations
1.6 Analysis and presentation of results
The rapid assessment report must be:
Clear Decision-makers or staff of local, national and interna-
tional organizations whose actions depend on the results
of the rapid assessment may not be trained in epidemio-
logy. User-friendly language and graphs make complex
data and trends easier to understand.
Standardized Results should be presented according to a standard
format so they can be compared with other assessments.
Action-oriented Clear recommendations should be made to implementing
and prioritized organizations, giving highest priority needs.
Widely distributed Copies of the report should be distributed to all organi-
zations involved in the emergency response.
Timely The assessment and report should be finalized and
distributed as quickly as possible, preferably within 3–4
days. Donors are often under political pressure in the first
few days after an emergency to demonstrate support by
their government and have access to funds. They must
have data to base their decisions on funding priorities.
32. 26 * Communicable disease control in emergencies – A field manual
1.6.1 Further reading
Assefa F et al. Malnutrition and mortality in Kohistan district, Afghanistan, April
2001. Journal of the American Medical Association, 2001, 286:2723–2728.
Gessner BD. Mortality rates, causes of death, and health status among displaced
and resident populations of Kabul, Afghanistan. Journal of the American Medical
Association, 1994, 272:382–385.
Epi Info. Centers for Diseases Control and Prevention, Atlanta, GA, USA. Available
from http//www.cdc.gov/epiinfo.
EpiData. The EpiData Association, Odense, Denmark. Available from http://www.
epidata.dk.
Henderson RH, Sundaresan T. Cluster sampling to assess vaccination coverage:
a review of experience with a simplified sampling method. Bulletin of the World
Health Organization, 1982, 60:253–260.
Lwanga SK, Lemeshow S. Sample size determination in health studies: a practical
manual. Geneva, World Health Organization, 1991.
Malilay J, Flanders WD, Brogan D. A modified cluster-sampling method for post-
disaster rapid assessment of needs. Bulletin of the World Health Organization,
1996, 74:399–405.
Porter JDH, van Look FL, Devaux A. Evaluation of two Kurdish refugee camps in
Iran, May 1991: the value of cluster sampling in producing priorities and policy.
Disasters, 1993, 17:341–347.
Rapid health assessment protocols for emergencies. Geneva, World Health
Organization, 1999.
Roberts L, Despines M. Mortality in eastern Democratic Republic of Congo. Lancet,
1999, 353(9171):2249–2250.
Roberts L. Mortality in eastern Democratic Republic of Congo: results from eleven
mortality surveys. Final draft. New York, International Rescue Committee, 2001.
http://intranet.theirc.org/docs/mortII_report_small.pdf
http://intranet.theirc.org/docs/mortII_graphs.pdf
http://intranet.theirc.org/docs/mortII_map.pdf
http://intranet.theirc.org/docs/mortII_exec.pdf
Rothenberg RB et al. Observations on the application EPI cluster survey methods
for estimating disease incidence. Bulletin of the World Health Organization, 1985,
63:93–99
33. 2. PREVENTION * 27
2. PREVENTION
This includes good site planning; provision of basic clinical services, shelter,
clean water and proper sanitation; mass vaccination against specific diseases;
a regular and sufficient food supply; and control of disease vectors. Table 2.1
lists the main diseases and disease groups targeted by such interventions.
Table 2.1 Diseases targeted by preventive measures
Preventive measure Impact on spread of:
Site planning diarrhoeal diseases, acute respiratory infections
Clean water diarrhoeal diseases, typhoid fever, guinea worm
Good sanitation diarrhoeal diseases, vector-borne diseases, scabies
Adequate nutrition tuberculosis, measles, acute respiratory infections
Vaccination measles, meningitis, yellow fever, Japanese encephalitis, diphtheria
Vector control malaria, plague, dengue, Japanese encephalitis, yellow fever, other
viral haemorrhagic fevers
Personal protection malaria, leishmaniasis
(insecticide-treated nets)
Personal hygiene louse-borne diseases: typhus, relapsing fever, trench fever
Health education sexually transmitted infections, HIV/AIDS, diarrhoeal diseases
Case management cholera, shigellosis, tuberculosis, acute respiratory infections, malaria,
dengue haemorrhagic fever, meningitis, typhus, relapsing fever
2.1 Shelter
In many emergency situations, the displaced population must be sheltered in
temporary settlements or camps. The selection of sites must be well planned
to avoid risk factors for communicable disease transmission, such as
overcrowding, poor hygiene, vector breeding sites and lack of adequate shelter.
Such conditions favour the transmission of diseases such as measles,
meningitis and cholera. Usually, the most suitable land is already occupied by
the local population, leaving less desirable areas available to refugees or
displaced people. Critical factors to consider when planning a site are: water
availability, means of transport, access to fuel, access to fertile soil and for
security reasons, a sufficient distance from national borders or frontlines.
The surrounding environment may also pose a threat to health in the form of
vectors not encountered in the population’s previous place of residence. In
order to reduce such risks it is essential that site selection, planning and organi-
zation be undertaken as soon as possible.
34. 28 * Communicable disease control in emergencies – A field manual
2.1.1 Site selection criteria
Settlements should avoid the major breeding sites of local vectors, as well as
marshy areas and flat, low-lying ground at risk of flooding. Preference should
be given to gently sloping, well drained sites on fertile soil with tree cover,
sheltered from strong winds. Local expertise and knowledge of the biology of
the vectors should be considered, such as avoiding forested hills in some Asian
countries where vectors proliferate. If not already sufficiently documented by
national and local health services, the epidemiological characteristics of the
area need to be assessed quickly.
The following criteria should be considered when assessing site suitability;
other criteria may also be relevant in specific situations.
Water supply
The availability of an adequate amount of safe water throughout the year has
proved in practice to be the single most important criterion for site location. The
water source should be close enough to avoid transporting water by trucks,
pumping it over long distances or walking long distances to collect insufficient
quantities.
Space
There must be enough space for the present number of emergency-affected
population, with provision for future influxes and for amenities such as water
and sanitation facilities, food distribution centres, storage sites, hospitals, clinics
and reception centres.
Topography and drainage
Gently sloping sites above the flood level is preferred in order to provide natural
drainage. Flat areas, depressions, swamp, river banks and lakeshore sites
should be avoided. Windy sites are unsuitable, as temporary shelters are
usually fragile.
Soil conditions
The soil type affects sanitation, water pipelines, road and building construction,
drainage and the living environment (in terms of dust and mud). The most
suitable soil type is one that will easily absorb human waste.
Access
The site should be accessible at all times (e.g. for food deliveries, roads during
rains).
Vegetation
The site area should have good vegetation cover if possible. Trees and plants
provide shade, help to prevent soil erosion, allow recharge of the groundwater
supplies and help in reducing dust. It may sometimes be necessary, however,
to destroy poisonous trees or plants, for example where populations are
accustomed to collecting berries or mushrooms.
35. 2. PREVENTION * 29
Environmental health
Areas near vector breeding sites where there is a risk of contracting malaria,
onchocerciasis (river blindness), schistosomiasis, trypanosomiasis, etc. should
be avoided.
Security
The site chosen should be in a safe area, sufficiently distant from national
borders and combat areas.
Local population
The use of land for a camp can cause friction with local farmers, herdsmen,
nomads and landowners. Some potential sites may have special ritual or
spiritual significance to local people, and site selection must respect the wishes
of the local population. Streams or rivers used for bathing and laundry may
cause pollution far downstream; water abstraction will reduce flow rates.
Indiscriminate defecation in the early stages may also pollute water supplies
used by the local population.
Fuel supply
Fuel for cooking is an essential daily requirement. Options for fuel include wood,
charcoal and kerosene. In practice, wood from surrounding forests is the most
likely fuel. It is important to liaise closely with the local forestry department to
control indiscriminate felling and collection.
2.1.2 Site layout and design
It is important to prepare a master plan of the camp. The site plan should be
sufficiently flexible to allow for a greater than expected influx of people. A 3–4%
per year population growth rate must also be planned for.
Overdevelopment of some areas of the site must be avoided as it can cause
health problems, especially for people who come from sparsely populated
environments.
Tribal, ethnic or religious differences may exist within the camp population or
between this population and the local people, or such groupings may develop or
be strengthened with time. The camp must be planned in such a way that these
divisions are honoured.
Site planning norms are presented in Table 2.2. The recommended figures for
camp layout and services are only guidelines. In severely overcrowded,
spontaneously settled camps it may be very difficult to achieve the
recommended figures during the initial emergency phase and realistic
compromises will have to be made. Nevertheless, the figures provide the basis
for planning and are the targets at which to aim.
36. 30 * Communicable disease control in emergencies – A field manual
Table 2.2 Site planning norms
Area per person for collective activities 30 m2 a
Shelter space per person 3.5 m2 b (4.5–5.5 m2 c in cold climates)
Distance between shelters 2 m minimum
Area for support services 7.5 m2/person
Number of people per water point 250
Number of people per latrine 20
Distance to water point 150 m maximum
Distance to latrine 30 m
Distance between water point and latrine 100 m
Firebreaks 75 m every 300 m
a
In practice this may be difficult to achieve, for example in areas with a high population density where little land
is available. This figure includes roads, services, shelter, etc. but depends on the layout and terrain. It does not
include land for livestock or agriculture. After space for covered shelter and support services, the remainder of
the 30 m2/person area is for family plot space, latrines, washing and cooking areas, community space, roads,
firebreaks, drainage, burial grounds and contingencies.
b
For a five-person family this equals a shelter 6 by 3 metres in a plot 15 by 10 metres.
c In cold climates where cooking is done indoors, extra shelter space is required.
2.1.3 Community participation
Ongoing community involvement in site planning and management is crucial
and can maximize the effectiveness of the intervention.
2.1.4 Location of family dwellings
The layout of dwellings relative to each other can have a significant impact on
security and cultural activities, and is important for the building of a social
structure. It also affects the use of latrines and water points. Although shelters
arranged in straight lines on a close grid pattern might appear to ease some
aspects of camp management, such a pattern is not normally conducive to
social cohesion. The camp should be organized into small community units or
“villages” each of approximately 1000 people. Traditional living patterns should
always be taken into account. Several villages can be combined to form a group;
several groups can form a section; and there can be several sections in one
camp. Table 2.3 shows the recommended structural organization for a camp
setting. Each group or section will require a number of decentralized services,
which are listed in Table 2.4.
The grouping of family plots into community units provides a defined, secure
space within each unit. People know each other and strangers will stand out. The
circumstances of an emergency may give rise to additional personal security
risks. Women may be vulnerable to harassment and rape. Ethnic and factional
37. 2. PREVENTION * 31
divisions can provoke violent confrontations. In these circumstances the
protection aspects of “shelter”’ may mean keeping different refugee groups apart
and/or the provision of secure compounds for particularly vulnerable refugees.
Table 2.3 Camp building blocks
1 family = 4–6 people
16 families = 80 people = 1 community
16 communities = 1250 people = 1 block
4 blocks = 5000 people = 1 sector
4 sectors = 20 000 people = 1 camp
2.1.5 Shelter design
A minimum shelter space of 3.5 m2 per person is recommended in emergency
situations. If possible, the emergency-affected population should build their
own shelters, preferably using local materials such as timber, grass, bamboo,
mud, sand and woven mats. Woven matting, natural fibre screens and bamboo
make very good ventilated walls. When necessary, rolled-up plastic sheeting
can be let down to make these walls water-, draught- and dust-proof. Tents and
plastic sheeting provide reasonable protection from the elements, but with
large numbers of people many units are required. Plastic sheeting may last
only 6–9 months, depending on the quality used; it degrades as a result of
exposure to the elements, especially sunlight. Canvas tents can last for up to
two years if well maintained. The build-up of dirt or rainwater on the roof, or
dirt on the walls, will shorten the life of a canvas tent.
It is best to plan the layout of shelter areas in community clusters adjacent to
the relevant latrines, water points and washing areas. These community units
should be as close as possible in design and layout to those with which the
population is most familiar.
2.1.6 Location of site services
Consideration must be given to the location of roads, houses, food and water
distribution points, emergency services (security, fire, medical), drains, washing
areas, latrines and solid waste pits. Public buildings require access roads for
vehicles and should be centrally located where possible.
Food distribution centres must be centrally located, with sufficient room for
crowds of people waiting and for trucks delivering food. Good design can help
considerably in crowd control and theft prevention. The main health facility
must be in a safe and accessible place, preferably on the periphery of the site
to allow for future expansion and to avoid overcrowding.
38. 32 * Communicable disease control in emergencies – A field manual
A site for a cholera treatment centre must be identified in advance, separate
from other health facilities and in an area where water supplies cannot be
contaminated.
Support facilities must be located away from dusty or potentially dangerous
major access roads.
Table 2.4 Main facilities on settlements
Centralized Decentralized
Administration Community health centres
Coordination offices Bathing and washing areas
Warehouse Social centres
Registration Schools
Hospital (for large camps) Recreation space
Tracing centres Supplementary feeding centres
Therapeutic feeding centres Religious buildings
Food distribution centres Water points
Training centres Latrines
Sanitation offices
Roads and firebreaks
Markets
2.1.7 Reception and registration
A reception area must be set up outside the settlement to receive and register
new arrivals before they become integrated within the camp. The registration
site should preferably be a large, flat, open space with a water supply and
latrines or defecation areas. Temporary first-night shelter and land for
accompanying animals may be needed.
2.1.8 Markets
Market areas are important trading and social centres, but they can pose health
risks where food and drink is for sale. The planning and layout of such areas
are very important. If possible the market should be outside the camp, or
several small market areas can be established. Vector control, waste collection
and disposal measures need to be particularly stringent at market areas.
Markets must be divided into food and non-food areas. Food areas should be
further divided into areas for raw and processed foods.
Areas must be provided for the slaughter of livestock, if possible with a concrete
slab with good drainage to carry away blood and animal droppings (although
one needs to ensure that this does not drain directly into a watercourse).
39. 2. PREVENTION * 33
2.1.9 Noise avoidance and traffic
Generators and pumps should be located away from family dwellings and the
buildings housing them should be soundproofed, with sufficient ventilation for
the escape of exhaust fumes. Traffic should be limited to main routes.
2.1.10 Camp coordination
Coordination between the various organizations working in the emergency is
essential in order to maximize positive impact on the population by means of
effective management and integration of relief activities.
The following steps are necessary to achieve this objective:
• establish clear leadership,
• create a coordinating body,
• ensure that programme activities are shared by agencies,
• clarify the roles and responsibilities of all partners,
• prevent duplication of activities,
• establish good communication channels,
• ensure that all needs are addressed,
• create and implement agreed common policies, standards and guidelines.
2.1.11 Liaison with local communities
Continued liaison with local communities is essential. The influx of emergency-
affected populations into their area means that they are now affected by the
emergency. There is a real risk of generating resentment if local people feel
that the emergency-affected populations are better served than they are. There
may be a need to provide medical or other assistance to local communities,
both to ensure equity and to prevent the spread of disease.
2.2 Water
Water and sanitation are vital elements in the transmission of communicable
diseases and in the spread of diseases prone to cause epidemics. Diarrhoeal
diseases are a major cause of morbidity and mortality among emergency-
affected populations, most being caused by a lack of safe water, inadequate
excreta disposal facilities and poor hygiene (see Table 2.5).
The goal of a water and sanitation programme is to minimize risks to the health
of a population, particularly one caught up in the difficult circumstances of an
emergency with its attendant displacement and dangers. Such a programme is
an integral part of preventive health activities.
The main focus of such a programme is on:
• the provision of a safe and sufficient water supply,
40. 34 * Communicable disease control in emergencies – A field manual
• provision for excreta disposal and the establishment of other waste control
and hygiene measures,
• a programme of public education for the affected population on the issues
of hygiene and water use.
Table 2.5 Water-related diseases
Diseases that occur owing to a lack of Skin infections: scabies, impetigo
water and poor personal hygiene Ophthalmic infections: conjunctivitis, trachoma
Louse-borne diseases: typhus, relapsing fever, trench fever
Diseases that occur owing to poor Caused by faecal pollution: cholera, typhoid, other diarrhoeal
biological quality of the water diseases, hepatitis A, hepatitis E, schistosomiasis
Caused by the urine of certain mammals: leptospirosis
Conditions that occur owing to poor Poisoning
chemical quality of the water
Diseases caused by water-based Malaria, dengue fever, onchocerciasis, yellow fever, Japanese
insect vectors encephalitis, guinea worm
In an emergency, the affected populations need immediate access to a water
supply in order to maintain health and to reduce the risk of epidemics. If the
emergency-affected population have to be sheltered in temporary settlements
or camps, water supply is an essential consideration in choosing the site
location. An adequate amount of safe drinking-water must be provided for the
entire displaced population.
The first objective is to provide sufficient water; quality can be addressed later.
Sufficient water of low quality is better than very little water of high quality.
During the rapid assessment of a proposed site it is essential to protect existing
water sources from possible contamination. If the population have already
moved into the area in question, then immediate measures should be taken to
isolate and protect the water source, if it is on or near the site.
Essential water requirements
• The minimum amount of water required in extreme situations is 7 litres per person
per day (only tolerable for a short duration). This amount does not reduce the
risk of epidemics in the population as it permits only a very low level of hygiene.
• The emergency requirement guideline is 20 litres per person per day. This allows
for cooking, laundry, bathing and activities essential to preventing the transmission
of water-borne diseases.
41. 2. PREVENTION * 35
2.2.1 Guidelines for assessing the volume of water required
The assessment of the volume of water required must take into account daily
population requirements and the effects of climate on the water source. The
effects on water requirements of a change in population size also need to be
estimated. A factor of 10–15% should be added to the total daily requirement of
a camp in order to provide for public institutions. Table 2.6 details the water
requirements in health facilities.
Table 2.6 Water requirements in health facilities
Hospital ward 50 litres/person per day
Surgery/maternity 100 litres/person per day
Dressing/consultation 5 litres per dressing
Feeding centre 20–30 litres/person per day
Kitchen 10 litres/person per day
When water is scarce, rationing should be introduced to ensure that the weak
and vulnerable survive and that an equitable distribution is achieved. In this
situation, monitoring is essential.
2.2.2 Providing a water supply
Identification of possible sources
The affected population must be involved in this process from the start, as they
will be relied on for repairs and maintenance in the future. All available sources
of water should be considered: a combination of sources may be used.
Assessment of water sources
Assessment is needed to determine:
• the quality of the water,
• the type of treatment needed,
• the method of extraction from the source,
• the most suitable distribution system.
Water quality
The key to disease prevention through water supply is ensuring that water is of
a high quality when consumed, not only just after treatment or at distribution
points. If people do not have enough water of acceptable quality, then they will
take water from other sources, which will most likely be contaminated.
42. 36 * Communicable disease control in emergencies – A field manual
In an emergency, biological quality is of greater importance than chemical
quality. The WHO guidelines detailed in Table 2.7 list the basic requirements
and parameters that must be measured. Groundwater sources usually yield
water of good quality, but chemicals that produce a bad odour or taste may also
be released into the water from underlying rocks.
Table 2.7 WHO guidelines on water quality
Criteria Guidelines
Faecal coliforms a < 10 per 100 ml
Odour/taste not detectable
Turbidity b < 5 NTU
Total dissolved solids c < 500 parts per million
pH d 6.5–8.5
a
Faecal coliforms are bacteria of faecal origin from the faeces of an animal (including humans). This parameter is
the most important when testing water for drinking. Fewer than 10 coliforms per 100 ml of water is acceptable.
The preferred level is zero, but this may not be practical in some cases.
b
Turbidity refers to water clarity. It is measured in “nephelometric turbidity units” (NTU). Turbidity may only be of
aesthetic importance, but this will matter to the affected population. It does inhibit the effectiveness of chlorine in
purifying the water and may also be an indication of the level of pollution.
c
Total dissolved solids refers to the quantity of dissolved matter in the water. It is measured in parts per million
(ppm). Drinking-water should have less than 500 ppm. Again, this relates to acceptability by the consumers.
d
pH is a measure of the acidity or alkalinity of the water. Alum (aluminium sulfate, used in the flocculation of
suspended solids) works more efficiently at a pH between 6 and 8. The pH should be less than 7 before adding
chlorine. Lime can be added to raise the pH and hydrochloric acid to lower it.
Water treatment
Water that does not meet the required standards must be treated before it is
distributed to the population. Table 2.8 presents the main methods of water
treatment, the selection of which depends on the extent and type of purification
required. Treatment is usually followed by chemical disinfection of the water,
the most common and effective disinfecting solution in emergency situations
being chlorine (see Table 2.9).
Quality control
Water quality checks must be made at regular intervals throughout all stages
of the water distribution chain. Tests are needed several times a day at the
beginning and in the middle of the chain for free residual chlorine. Weekly
checks for faecal coliforms are needed in emergency situations, and particularly
during epidemics.
43. 2. PREVENTION * 37
Distribution
If the rapid assessment indicates that a suitable water system will take some
time to develop, short-term measures such as trucking may have to be
considered. In such instances, rationing may be necessary to ensure equal
distribution among the entire population.
Once a satisfactory supply has been established, enough storage and reserve
systems must be developed to allow for maintenance and breakdowns in supply
and equipment. Storage of at least one day’s requirement must be provided.
Preventing contamination from sanitation facilities and other sources of
pollution is of paramount importance. The location, organization and
maintenance of water points is detailed in Table 2.10.
Table 2.8 Methods of water treatment
Storage This is the simplest method of improving water quality. If water is stored in a covered
tank for a period of time, pathogenic bacteria die off and sink to the bottom by a
simple sedimentation process. Two days is the minimum length of storage
recommended. The water will not necessarily be totally free of contamination by
simple filtration.
Storage tanks require cleaning and de-sludging at regular intervals, depending on
the level of sediment in the water. Algal build-up should be prevented. No animals
or unauthorized persons should be allowed access to the tanks.
Aeration Aeration is achieved by allowing the water to cascade over layers of gravel. Aeration
may be required if iron or manganese is present in the water, since these give an
unpleasant taste and a brownish discoloration to food and clothes.
Sedimentation Water from river sources, especially in the rainy season, often has a high silt content.
Simple storage methods are not sufficient for this silt to settle. Along with the natural
sedimentation process, the addition of a chemical coagulant, usually aluminium
sulfate (alum), is necessary.
The amount of alum needed depends on the amount of suspended matter in the
water, the turbidity, the pH and the hardness of the water. Effluent water should not
contain a concentration of alum greater than the guideline figure given in the WHO
guidelines for drinking-water quality.
Filtration Slow and rapid sand filters.
Disinfection Chlorine is the most common and effective disinfecting solution in emergency
situations, and various dilutions are used in different situations (see Table 10). The
amount of chlorine required depends on the quantity of organic matter and of harmful
organisms in the water.
The dose should leave a residual level of chlorine of between 0.2 and 0.5 mg/litre
(a higher level will leave a taste and people will not drink the water). A simple drip-
feed tank can be designed to administer the correct amount of chlorine.